From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id BC36C9E7 for ; Tue, 14 Feb 2017 18:30:18 +0100 (CET) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by fmsmga101.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 14 Feb 2017 09:30:17 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.35,162,1484035200"; d="scan'208";a="1094707851" Received: from irsmsx109.ger.corp.intel.com ([163.33.3.23]) by orsmga001.jf.intel.com with ESMTP; 14 Feb 2017 09:30:16 -0800 Received: from irsmsx103.ger.corp.intel.com ([169.254.3.77]) by IRSMSX109.ger.corp.intel.com ([169.254.13.44]) with mapi id 14.03.0248.002; Tue, 14 Feb 2017 17:30:14 +0000 From: "Richardson, Bruce" To: "Yigit, Ferruh" CC: Thomas Monjalon , "dev@dpdk.org" Thread-Topic: [dpdk-dev] [RFC 17.05] test: move tests to separate folder Thread-Index: AQHShtT4Fs5/wL0tWUeP6qO4in6AKqFoo/aAgAAX8wCAAAVDgA== Date: Tue, 14 Feb 2017 17:30:14 +0000 Message-ID: <59AF69C657FD0841A61C55336867B5B035B9815F@IRSMSX103.ger.corp.intel.com> References: <20170214151326.7554-1-ferruh.yigit@intel.com> <20170214154131.GA20524@bricha3-MOBL3.ger.corp.intel.com> <8e969f23-6a65-553e-25ca-593c01c9fe8b@intel.com> In-Reply-To: <8e969f23-6a65-553e-25ca-593c01c9fe8b@intel.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-titus-metadata-40: eyJDYXRlZ29yeUxhYmVscyI6IiIsIk1ldGFkYXRhIjp7Im5zIjoiaHR0cDpcL1wvd3d3LnRpdHVzLmNvbVwvbnNcL0ludGVsMyIsImlkIjoiNjZjNzIwOWYtOGJhYi00YzgxLTk3YTktMWE3NGQ4OWIyNjVkIiwicHJvcHMiOlt7Im4iOiJDVFBDbGFzc2lmaWNhdGlvbiIsInZhbHMiOlt7InZhbHVlIjoiQ1RQX0lDIn1dfV19LCJTdWJqZWN0TGFiZWxzIjpbXSwiVE1DVmVyc2lvbiI6IjE2LjIuMTEuMCIsIlRydXN0ZWRMYWJlbEhhc2giOiJXdGY0TlZXNVI3OElmTXFVUVFMWGFYSFVMMU13clBIVmYwRTk0Ujk4TkhBPSJ9 x-ctpclassification: CTP_IC x-originating-ip: [163.33.239.181] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [RFC 17.05] test: move tests to separate folder X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Feb 2017 17:30:19 -0000 > -----Original Message----- > From: Yigit, Ferruh > Sent: Tuesday, February 14, 2017 5:07 PM > To: Richardson, Bruce > Cc: Thomas Monjalon ; dev@dpdk.org > Subject: Re: [dpdk-dev] [RFC 17.05] test: move tests to separate folder >=20 > On 2/14/2017 3:41 PM, Bruce Richardson wrote: > > On Tue, Feb 14, 2017 at 03:13:26PM +0000, Ferruh Yigit wrote: > >> Tests are part of app folder and compiled with library every time. > >> Moving tests into a "test" folder which won't be compiled by default. > >> To compile tests, need to give explicit "make test" command. > >> > >> "make test" was previously used to run tests, which renamed to "make > >> test_run" with this patch. > >> > >> This makes default compilation ~30% faster, > >> [clang, make -j8, old]: real 1m04.355s > >> [clang, make -j8, new]: real 0m41.740s > >> > >> For new case, test needs to built separately, which takes, > >> [clang, make -j8 test]: real 0m24.293s > >> > >> The point is tests are not required always and by every one. > >> > >> Signed-off-by: Ferruh Yigit > >> --- > > > > Hi Ferruh, > > > > I'm not sure I'm convinced by this, as I think there are advantages to > > having the test code always compiled. Anything that is not compiled in > > DPDK by default is more likely to be broken by patch submissions. The > > speed boost to build is nice, but I'm not sure it's worth it. > > However, I'm open to being convinced otherwise on this... >=20 > Perhaps I should send this as two patches, first separate the unit tests > and second disable tests in default compilation. >=20 > Your concern is about disabling test compilation by default. >=20 > - If nobody interested in running test, why force them to compile. If > people interested in unit tests, test will be compiled and issues will be > resolved. > Unit tests are for developers more than end users, and if developers are > not using unit tests we may have another thing to focus. >=20 > - This should be automated somehow, unit test should be build and run > regularly and automatically. So we can see when it is broken. >=20 Ok, makes sense. Just to be awkward :-), one last question: Why separate building and runnin= g the tests? My suggestion would be to have "make test" both build and run th= e tests. If there is no work to do in building them, then the time cost of th= e extra empty build is negligible compared to the time taken to actually run the tests. If we really want to build the tests without running them I'd suggest "build_test" target instead. This keeps "make test" pretty much as it was before. /Bruce